Nate Cohn Profile picture
15 Oct, 10 tweets, 2 min read
Lindsey Graham holds modest lead in the race for U.S. Senate in South Carolina, according a new Times/Siena poll.
Graham leads Harrison, 46 to 40.
Bill Bledsoe--who has dropped out--holds 4 percent, and another 2 percent won't vote for Senate.
nytimes.com/2020/10/15/us/…
Notably, Graham quite a bit better in interviews over the last two nights during the confirmation hearings.
He ran well behind the president over the weekend, but ran well ahead of the president over the last two days. Maybe noise. Maybe not. It'll be interesting to see.
In the presidential race, Donald Trump holds a slightly larger 49 to 41 percent lead over Joe Biden.
Unlike some other polling out there, we find that Republicans continue to hold an overwhelming advantage among white voters without a degree in the state, who back Trump by a 77-18 margin.
In terms of the composition of the electorate, the sample is 71% white and 26% black, based on how voters described their race when they registered to vote. That represents an uptick in the black share of the electorate from '16, and there's no RV-LV gap here
Yeah, so there's an interesting gap here--how voters respond to our poll question about race and ethnicity, and how they responded to the race question on their voter registration form.
Our sample is 65-23 white/black by self-report in the poll. When those same folks registered to vote, they were 71-26 white/black. We weight on the latter definition, since we know exactly how many people identified as white or black on their voter registration form
If you crosstab the results on *voter file* race, instead of self-report, you get the same thing:
Black: 82-3 in the presidential race
White 26-66.
And I see a lot of talk about undecided voters, but like, if you just allocate the undecided vote by Trump approval you get up to... Graham +7.6. If you allocate with the presidential race ballot, you get Graham+7.3.
I think it's a lot better than eye-balling the pct. undecided by demographic group in the crosstabs

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Nate Cohn

Nate Cohn Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @Nate_Cohn

16 Oct
Republicans lead in Alaska, according to a new Times/Siena poll, as voters sour on the president but hesitate to embrace Dems.
--
Trump leads Biden, 45 to 39. Jorgensen (L) at 8.
Sullivan leads Gross, 45 to 37. Howe (AIP) at 10.
Trump approval 47/47
nytimes.com/2020/10/16/ups…
In the US House rematch, Don Young leads Alyse Galvin 49-41, about the same as the outcome two years ago.
What to make of those third party tallies? I mean, I'd sure guess a lot of it will drop off at the ballot box. But Alaska does have a reputation for backing minor party candidates. Looking at their attitudes about Trump, it's not super obvious that anyone's poised to benefit
Read 9 tweets
16 Oct
As you probably know, we're going to have an Alaska poll out at 1PM today. It's a notoriously tough state and this was our first time polling it. Long time listeners know that sometimes our first time polling a state can be messy! I didn't think this one fell in that category
TBH, I'd rather poll Alaska than Michigan. And if we had to do it again, I think we did learn a little bit to help tighten it up even more--though I'm comfortable with what we've got.
On cautionary step we took: we only went for N=425. We just didn't know whether rural AK was just tough like totally absolutely horrible. We didn't want to be in a position where we literally called every number we had out there without enough responses.
Read 8 tweets
15 Oct
One seemingly tedious task that I've come to enjoy is recoding the race of our respondents who say they're of 'some other race,' and then offer an answer. You never know what you're going to get
There are constants: Irish and Italian-Americans identifying as such. A handful of racists: "Aryan." Ridiculous answers: "Ginger," "a big fat gay mulatto!" Interesting detective work, where an interviewer types an obscure answer phonetically--"Chupic"--and you figure it out
('Chupic" appeared to be 'Cup'ik' an Alaska Native tribe, and the respondent was in the geographic part of western Alaska for it, which was fun)
Read 8 tweets
14 Oct
In general, I'm pretty open-minded about the range of possibilities in this election. I'm also very slow to judge pollsters based on crosstabs, since I think so much of what goes on down there is just noise that cancels out. But...
Quinnipiac has had *a lot* of polls showing Biden doing really, really well among white southerners, including white southerners without a degree in the Deep South. And hey, maybe it'll happen. But that's a far-fetched possibility IMO, and pollster bias is the easier explanation
And no, I don't know why or how. I mean, even our sample of white no degree voters in ATL+inner burbs was just 30% Biden--I assume they call outside of those four counties. Biden was at 13% in the rest of the state
Read 4 tweets
14 Oct
I honestly don't understand why people want to push undecided voters to make up a mind in a race when they barely know the candidates. Knowing that they're undecided is a lot more valuable
There are some structural reasons why we do get higher undecided voters than other polls:
--probablistic LV screen keeps some number of people who say they won't vote, who are far likelier to be undecided
--we have far more low turnout voters, who are far likelier to be undecided
--we're naming the third party candidates, and in any given survey a decent number are backing a minor party candidate, even if I don't tweet about them
Read 5 tweets
13 Oct
Biden's lead is up to 12.7 on the USC tracker, his all time high.
election.usc.edu
It's a particularly interesting day to take note of the USC tracker, since it's a two week field-period and it's been two long weeks since the first presidential debate. So today will be the last day, I believe, with any pre-debate data.
Biden's lead has grown 3.2 pts since 9/29
Biden's gains ought to begin to slow once all of the pre-debate interviews are cycled out. It could even begin to decline, if more recent interviews are better for Trump than immediately post-debate. So it'll be a relatively interesting few days for following the USC trends
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!